CN117037432B - Risk evaluation geological disaster early warning method based on multi-method cooperation - Google Patents
Risk evaluation geological disaster early warning method based on multi-method cooperation Download PDFInfo
- Publication number
- CN117037432B CN117037432B CN202311288116.5A CN202311288116A CN117037432B CN 117037432 B CN117037432 B CN 117037432B CN 202311288116 A CN202311288116 A CN 202311288116A CN 117037432 B CN117037432 B CN 117037432B
- Authority
- CN
- China
- Prior art keywords
- data
- risk
- prediction model
- model
- geological
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000011156 evaluation Methods 0.000 title claims abstract description 40
- 238000012544 monitoring process Methods 0.000 claims abstract description 59
- 238000012502 risk assessment Methods 0.000 claims abstract description 52
- 238000005516 engineering process Methods 0.000 claims abstract description 34
- 238000012549 training Methods 0.000 claims abstract description 27
- 230000007246 mechanism Effects 0.000 claims abstract description 18
- 238000010801 machine learning Methods 0.000 claims abstract description 12
- 238000012360 testing method Methods 0.000 claims description 19
- 238000007726 management method Methods 0.000 claims description 14
- 238000012417 linear regression Methods 0.000 claims description 13
- 239000002689 soil Substances 0.000 claims description 12
- 238000003066 decision tree Methods 0.000 claims description 10
- 230000002265 prevention Effects 0.000 claims description 10
- 238000006073 displacement reaction Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 8
- 238000011835 investigation Methods 0.000 claims description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 6
- 239000003673 groundwater Substances 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 5
- 230000006872 improvement Effects 0.000 claims description 5
- 238000005259 measurement Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 4
- 238000013480 data collection Methods 0.000 claims description 4
- 238000005065 mining Methods 0.000 claims description 4
- 238000005527 soil sampling Methods 0.000 claims description 4
- 238000007619 statistical method Methods 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 3
- 230000010354 integration Effects 0.000 claims description 3
- 230000035699 permeability Effects 0.000 claims description 3
- 238000002360 preparation method Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 2
- 230000008901 benefit Effects 0.000 description 8
- 230000008859 change Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 230000010485 coping Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000012876 topography Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/10—Alarms for ensuring the safety of persons responsive to calamitous events, e.g. tornados or earthquakes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
- G16Y40/00—IoT characterised by the purpose of the information processing
- G16Y40/10—Detection; Monitoring
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
- G16Y40/00—IoT characterised by the purpose of the information processing
- G16Y40/20—Analytics; Diagnosis
Abstract
The invention relates to the technical field of geological disaster early warning methods, in particular to a risk evaluation geological disaster early warning method based on multi-method cooperation. In the invention, the comprehensive influence among a plurality of parameters is comprehensively considered by collecting various parameter data related to geological disasters. And constructing a prediction model by utilizing machine learning, training and learning the collected parameter data, monitoring in real time by utilizing the Internet of things technology and a sensor, feeding the collected data back to the prediction model, and updating the risk assessment in real time. And grading the risks of the geological disasters so as to realize effective management of areas with different risk grades. And the new collected data is used for comprehensively evaluating and optimizing the prediction model at regular intervals so as to improve the accuracy and reliability of prediction. Feedback from various aspects is collected to further improve the performance of the predictive model by establishing a mechanism for public participation and information sharing.
Description
Technical Field
The invention relates to the technical field of geological disaster early warning methods, in particular to a risk evaluation geological disaster early warning method based on multi-method cooperation.
Background
The geological disaster early warning method is characterized in that by monitoring, analyzing and judging the development trend and the dangerous degree of the geological disaster, measures are taken in advance, and the possible geological disaster is warned in advance so as to reduce disaster loss. The system is an important component of the disaster prevention and reduction work of geological disasters, can provide precious time windows, enables related departments and public to take necessary emergency measures, reduces casualties and property loss, and has important significance for the disaster reduction work due to the fact that different types of geological disasters possibly need different early warning methods and technical means and the accuracy and timeliness of early warning.
In the actual use process of the existing geological disaster early warning method, the existing method is often based on a single parameter index, so that the comprehensive influence among a plurality of parameters is ignored, and the accuracy and reliability of early warning information are reduced. Secondly, the existing method has limitation on a prediction model, and the occurrence of geological disasters is difficult to predict accurately. Meanwhile, the existing method is limited in consideration of diversity and complexity of scenes, and is difficult to adapt to different geological disaster types and environmental conditions. In addition, the existing early warning method has complexity in terms of data processing and analysis and lacks of social participation and sharing mechanisms due to the lack of comprehensive performance evaluation and satisfaction of personalized requirements.
Disclosure of Invention
The invention aims to solve the defects in the prior art, and provides a risk evaluation geological disaster early warning method based on multi-method cooperation.
In order to achieve the above purpose, the present invention adopts the following technical scheme: a risk evaluation geological disaster early warning method based on multi-method cooperation comprises the following steps:
s1: collecting various parameter data related to a geological disaster;
s2: based on the collected parameter data, constructing a prediction model by utilizing machine learning, and identifying risk factors possibly causing geological disasters;
s3: real-time monitoring is carried out by utilizing the Internet of things technology and a sensor, real-time monitoring data are fed back to the prediction model, and risk assessment data are updated in real time;
s4: based on the risk assessment data, the risk grade of the geological disaster is divided into three grades, namely a high grade, a medium grade and a low grade, so that effective management of areas with different risk grades is realized;
s5: after the predictive model operates for a specified time period, comprehensively evaluating and optimizing the model by utilizing the newly collected data;
s6: feedback from various aspects is collected by establishing a mechanism of public participation and information sharing to further improve the accuracy of the predictive model.
As a further aspect of the present invention, the step of collecting various parameter data related to a geological disaster specifically includes:
s101: obtaining geological structure parameter data, including stratum structures, lithology distribution and fracture zones, through geological exploration and geological investigation;
s102: the method comprises the steps of obtaining topographic and geomorphic parameter data, including ground surface elevation, gradient, slope direction and topographic curvature, by utilizing a remote sensing technology and ground measurement;
s103: collecting weather parameter data including rainfall, snowfall, air temperature, humidity and wind speed through a weather observation station, weather satellites and weather radars;
s104: soil parameter data including soil type, water content and permeability are obtained through soil sampling and laboratory testing;
s105: collecting hydrological parameter data including groundwater level, river level and flood history through a water level measuring instrument and a hydrological observation station;
s106: seismic monitoring data are acquired through a seismic observation station, ground temperature data are acquired through a ground temperature measuring device, and slope displacement data are acquired through slope monitoring equipment.
As a further aspect of the present invention, the step of constructing the prediction model using machine learning based on the collected parameter data specifically includes:
s201: based on the collected parameter data, selecting characteristics, and selecting a participation data set;
s202: dividing the participation data set into a training set and a testing set;
s203: selecting a multiple linear regression combined decision tree to construct the prediction model;
s204: based on the training set, training and parameter tuning are carried out on the prediction model, and a trained prediction model is obtained;
s205: and evaluating the trained prediction model based on the test set.
As a further scheme of the invention, the feature selection specifically means that pearson correlation coefficients between each feature and geological disasters are calculated, the absolute value of each pearson correlation coefficient is close to 1, the linear relation between two variables is stronger, the parameter data are arranged in a descending order based on the pearson correlation coefficients, and 30% of the data located at the upper layer are selected as the participation data set;
the participation data set is divided into a training set and a test set, wherein 70% of data in the participation data set is used as the training set for training and parameter tuning, and the rest 30% of data is used as the test set for evaluating the performance and generalization capability of the model;
the multiple linear regression specifically refers to establishing a linear regression model by using the relation between the characteristics of the selected participation data set and the geological disasters, and describing the quantitative relation between the parameters and the geological disasters;
the decision tree specifically refers to layering division among a plurality of parameters based on a characteristic threshold value or information gain, and a geological disaster occurrence rule and risk factors are found.
As a further scheme of the present invention, the real-time monitoring is performed by using the internet of things technology and a sensor, the real-time monitoring data is fed back to the prediction model, and the step of updating the risk assessment data in real time specifically includes:
s301: arranging a sensor network in a detection area, and connecting the sensor with the network through the technology of the Internet of things, wherein the sensor comprises an earthquake sensor, a displacement sensor and a pressure sensor;
s302: the sensor collects monitoring data in real time and transmits the monitoring data to a data center through the internet of things technology;
s303: the data center feeds the monitoring data back to the prediction model based on a stream processing technology;
s304: and updating the prediction model by using real-time monitoring data, carrying out real-time risk assessment according to the output of the prediction model, and updating the risk assessment data.
As a further aspect of the present invention, the step of classifying the risk class of the geological disaster into three classes of high, medium and low based on the risk assessment data specifically includes:
s401: confirming risk indexes including terrain gradient and elevation, geological conditions, rainfall and rainfall intensity, temperature and meteorological conditions, groundwater level and hydrologic conditions, soil type and stability and geological disasters;
s402: weighting each risk index by using an analytic hierarchy process to obtain a weighted result;
s403: based on the risk index and the weighted result, establishing a risk assessment model by adopting a Bayesian network;
s404: the risk assessment data are imported into a risk assessment model, and the first 10% is defined as high risk, the middle 25% as medium risk and the last 65% as low risk according to the output result of the risk assessment model;
s405: and (5) according to regions with different risk grades, corresponding management measures and prevention strategies are formulated.
As a further scheme of the invention, corresponding management measures and prevention strategies are formulated according to regions with different risk levels, specifically, the high-risk regions adopt limited construction, enhanced monitoring and early warning systems;
taking earthquake preparation measures in the medium risk areas, including disaster education and exercise;
low risk areas focus on the trend and monitoring of risk.
As a further aspect of the present invention, the steps of performing comprehensive evaluation and optimization of the model using the newly collected data specifically include:
s501: storing the real-time monitoring data and the subsequent disaster event data as an updated data set;
s502: applying the updated data set to the prediction model that has been run to obtain a new prediction result;
s503: comparing and verifying the new prediction result with the actual observation value, and adopting root mean square error as an evaluation index to evaluate the accuracy and stability of the prediction model to obtain a comprehensive evaluation result;
s504: and optimizing the prediction model based on the comprehensive evaluation result.
As a further aspect of the present invention, the step of optimizing the prediction model based on the comprehensive evaluation result specifically includes:
adjusting parameters of the prediction model, and searching for an optimal parameter combination by using grid searching and random searching technologies so as to minimize a prediction error of the prediction model on an updated data set;
performing feature engineering on the updated data set, optimizing feature representation by adopting feature selection, feature transformation and feature combination, and improving the prediction capacity of the prediction model on geological disasters to serve as an optimized prediction model;
and carrying out a new round of risk assessment and prediction based on the optimized prediction model, comparing with an actual observed value, and further iterating and optimizing the model.
As a further scheme of the invention, the steps of collecting feedback of all aspects by establishing a public participation and information sharing mechanism to further improve the accuracy of the prediction model are specifically as follows:
s601: establishing an online platform, and setting up an online investigation tool as a data collection and integration mechanism;
s602: the public provides feedback opinion and knowledge through the online survey tool;
s603: integrating and summarizing based on the feedback opinion and knowledge, cleaning data, removing abnormal values and repeated data, and obtaining a feedback data set;
s604: based on the feedback data set, using a statistical method and a text mining technique to find patterns, trends and key problems in the feedback data set as available data;
s605: based on the improvement of the available data applied to the prediction model, the data set is updated and corrected for feedback of model input data to reflect actual conditions, and optimization and adjustment are performed according to analysis results for feedback of model parameters or algorithms.
Compared with the prior art, the invention has the advantages and positive effects that:
in the invention, the comprehensive influence among a plurality of parameters is comprehensively considered by collecting various parameter data related to geological disasters. A predictive model is constructed using machine learning by training and learning the collected parameter data to identify risk factors that may cause geological disasters. Real-time monitoring is performed by utilizing the Internet of things technology and a sensor, collected data is fed back to a prediction model, risk assessment is updated in real time, and early warning is timely sent out. And grading the risks of the geological disasters so as to realize effective management of areas with different risk grades. And the new collected data is used for comprehensively evaluating and optimizing the prediction model at regular intervals so as to improve the accuracy and reliability of prediction. Feedback from various aspects is collected to further improve the performance of the predictive model by establishing a mechanism for public participation and information sharing.
Drawings
Fig. 1 is a schematic diagram of main steps of a risk evaluation geological disaster early warning method based on multi-method cooperation;
fig. 2 is an S1 detailed schematic diagram of a risk evaluation geological disaster early warning method based on multi-method cooperation;
fig. 3 is an S2 detailed schematic diagram of a risk evaluation geological disaster early warning method based on multi-method cooperation;
fig. 4 is an S3 detailed schematic diagram of a risk evaluation geological disaster early warning method based on multi-method cooperation;
fig. 5 is an S4 detailed schematic diagram of a risk evaluation geological disaster early warning method based on multi-method cooperation;
fig. 6 is an S5 detailed schematic diagram of a risk evaluation geological disaster early warning method based on multi-method cooperation;
fig. 7 is an S6 detailed schematic diagram of a risk evaluation geological disaster early warning method based on multi-method cooperation.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
In the description of the present invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like indicate orientations or positional relationships based on the orientation or positional relationships shown in the drawings, merely to facilitate describing the present invention and simplify the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and therefore should not be construed as limiting the present invention. Furthermore, in the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
Embodiment one: referring to fig. 1, the present invention provides a technical solution: a risk evaluation geological disaster early warning method based on multi-method cooperation comprises the following steps:
s1: collecting various parameter data related to a geological disaster;
s2: based on the collected parameter data, constructing a prediction model by utilizing machine learning, and identifying risk factors possibly causing geological disasters;
s3: real-time monitoring is carried out by utilizing the Internet of things technology and a sensor, real-time monitoring data are fed back to the prediction model, and risk assessment data are updated in real time;
s4: based on the risk assessment data, the risk grade of the geological disaster is divided into three grades, namely a high grade, a medium grade and a low grade, so that effective management of areas with different risk grades is realized;
s5: after the predictive model operates for a specified time period, comprehensively evaluating and optimizing the model by utilizing the newly collected data;
s6: feedback from various aspects is collected by establishing a mechanism of public participation and information sharing to further improve the accuracy of the predictive model.
By collecting various parameter data related to geological disasters and constructing predictive models using machine learning, data-driven risk assessment can be performed. And the risk assessment data is updated in real time by combining the internet of things technology and real-time monitoring of the sensor, so that real-time geological disaster early warning is realized. Based on the risk assessment data, the risk grades of the geological disasters are divided into three grades, namely high, medium and low, and corresponding management and countermeasure are provided for areas with different risk grades. And model comprehensive evaluation and optimization are carried out regularly, and the prediction model is continuously improved by utilizing the newly collected data. And by establishing a public participation and information sharing mechanism, feedback in all aspects is collected, and the accuracy and the acceptability of the prediction model are improved. The implementation steps can improve the reliability and timeliness of the geological disaster risk assessment, provide effective support for decision makers, related departments and the public, and promote effective management of geological disaster risks and formulation of coping strategies.
The data-driven assessment method can improve the objectivity and accuracy of assessment, and provide scientific basis and accurate risk information for decision makers. The real-time monitoring and early warning system can timely sense the change and abnormal condition of geological disasters and provide accurate early warning information, so that related departments and the public can take effective precaution and emergency measures. The risk classification and management can help a decision maker to purposefully formulate management and resource allocation strategies, and the disaster coping capability of the society is improved. Periodic model evaluation and optimization can continuously improve the accuracy and stability of the prediction model, and enhance the practicability and reliability of the prediction model. And the public participation and information sharing mechanism can promote the interaction and cooperation of the public and decision makers, enhance the public safety awareness and increase the trust and acceptance of the public to the prediction model. Considering these effects comprehensively, the implementation of geological disaster risk assessment can effectively reduce the risk of geological disasters, improve the safety level of society, provide accurate and reliable information for decision makers, strengthen the participation awareness of the public, and provide important support for disaster management and social development.
Referring to fig. 2, the steps of collecting various parameter data related to geological disasters are specifically:
s101: obtaining geological structure parameter data, including stratum structures, lithology distribution and fracture zones, through geological exploration and geological investigation;
s102: the method comprises the steps of obtaining topographic and geomorphic parameter data, including ground surface elevation, gradient, slope direction and topographic curvature, by utilizing a remote sensing technology and ground measurement;
s103: collecting weather parameter data including rainfall, snowfall, air temperature, humidity and wind speed through a weather observation station, weather satellites and weather radars;
s104: soil parameter data including soil type, water content and permeability are obtained through soil sampling and laboratory testing;
s105: collecting hydrological parameter data including groundwater level, river level and flood history through a water level measuring instrument and a hydrological observation station;
s106: seismic monitoring data are acquired through a seismic observation station, ground temperature data are acquired through a ground temperature measuring device, and slope displacement data are acquired through slope monitoring equipment.
The collection and analysis of these various parameter data can be used to comprehensively assess the risk and potential risk of geological disasters. The geologic structure parameter data and the topography parameter data provide a basis for understanding the geologic hazard formation mechanisms and hazard areas. The meteorological parameter data and the hydrological parameter data are helpful for predicting and early warning meteorological hydrologic disasters such as flood, debris flow, waterlogging and the like. The soil parameter data provides stability and disaster resistance information of the soil, and helps to plan and manage land utilization. The geothermal data and the slope displacement data can monitor the signs and potential risks of geological disasters.
Referring to fig. 3, the steps of constructing a prediction model using machine learning based on the collected parameter data are specifically:
s201: based on the collected parameter data, performing feature selection, and selecting a participation data set;
s202: dividing the participation data set into a training set and a test set;
s203: selecting a multiple linear regression combined decision tree to construct a prediction model;
s204: based on the training set, training and parameter tuning are carried out on the prediction model, and a trained prediction model is obtained;
s205: based on the test set, the trained predictive model is evaluated.
Referring to fig. 3, feature selection specifically refers to calculating pearson correlation coefficients between each feature and a geological disaster, wherein the absolute value of each pearson correlation coefficient is closer to 1, which means that the linear relationship between two variables is stronger, parameter data are arranged in a descending order based on the pearson correlation coefficients, and data positioned at the upper 30% are selected as a participation data set;
the participation data set is divided into a training set and a test set, wherein 70% of data in the participation data set is used as the training set for training and parameter tuning, and the rest 30% of data is used as the test set for evaluating the performance and generalization capability of the model;
the multiple linear regression specifically refers to the establishment of a linear regression model by using the relation between the characteristics of the selected participation data set and the geological disasters, and describes the quantitative relation between the parameters and the geological disasters;
the decision tree specifically refers to that layering division among a plurality of parameters is performed based on a threshold value or information gain of the characteristics, and rules and risk factors of occurrence of geological disasters are found.
Based on the collected parameter data, the step of constructing a prediction model by utilizing machine learning comprises feature selection, data set division, model selection and construction, model training and parameter tuning and model evaluation. Specifically, the pearson correlation coefficient between each feature and the geological disaster is calculated through a feature selection method, and the first 30% with higher correlation is selected as the feature of the participation data set. The participation dataset was divided into training and test sets, with 70% of the data used for training and parameter tuning and the remaining 30% used for evaluating model performance and generalization ability. Multiple linear regression is selected in combination with decision trees to construct a predictive model, wherein the multiple linear regression describes the linear relationship of parameters to geological disasters, and the decision trees reveal nonlinear relationships between parameters and risk factors. And obtaining a trained prediction model by carrying out model training and parameter tuning on the training set. Finally, the trained model is evaluated by using a test set, and the fitting degree and accuracy of the model are measured by using evaluation indexes such as mean square error and decision coefficients. From the implementation perspective, the steps are beneficial to improving the accuracy of a prediction model, finding geological disaster rules and factors, optimizing the performance of the model and improving the generalization capability and applicability of the model. Such a machine learning predictive model construction process would provide powerful support for risk assessment and decision making of geological disasters.
Referring to fig. 4, the real-time monitoring is performed by using the internet of things technology and a sensor, real-time monitoring data is fed back to the prediction model, and the step of updating the risk assessment data in real time specifically includes:
s301: arranging a sensor network in a detected area, and connecting the sensor with the network through the technology of the Internet of things, wherein the sensor comprises an earthquake sensor, a displacement sensor and a pressure sensor;
s302: the sensor collects monitoring data in real time and transmits the monitoring data to the data center through the internet of things technology;
s303: the data center feeds the monitoring data back to the prediction model based on the stream processing technology;
s304: the prediction model is updated by utilizing the real-time monitoring data, and real-time risk assessment is carried out according to the output of the prediction model, so that the risk assessment data is updated.
The method for real-time monitoring by utilizing the Internet of things technology and the sensor and feeding the real-time monitoring data back to the prediction model comprises the following steps: and arranging sensor networks such as earthquake sensors, displacement sensors, pressure sensors and the like in a monitoring area, and connecting the sensors with the network through the technology of the Internet of things. The sensor collects monitoring data in real time and transmits the data to the data center through the internet of things technology. The data center feeds the monitoring data back to the prediction model by using a stream processing technology, so that the model is updated in real time. And carrying out real-time risk assessment and updating risk assessment data according to the output of the prediction model. The implementation scheme provides real-time monitoring capability, and early warning and finding out geological disaster risks in time; the data collection efficiency and accuracy are improved, the data is transmitted and stored in a centralized manner through the Internet of things technology, and the inconvenience of manual collection is avoided; realizing the real-time update of the prediction model so as to adapt to the real-time monitoring data and the dynamic change of geological disaster risks; and through real-time risk assessment, the change situation of geological disaster risks is mastered in time, and real-time early warning and decision support are provided for a decision maker. In conclusion, the internet of things technology and the sensors are utilized to monitor in real time and feed data back to the prediction model, so that real-time evaluation and decision support of geological disaster risks can be realized, and the efficiency and response speed of disaster management are improved.
Referring to fig. 5, the steps of classifying the risk level of the geological disaster into three levels of high, medium and low based on the risk assessment data are specifically as follows:
s401: confirming risk indexes including terrain gradient and elevation, geological conditions, rainfall and rainfall intensity, temperature and meteorological conditions, groundwater level and hydrologic conditions, soil type and stability and geological disasters;
s402: weighting each risk index by using an analytic hierarchy process to obtain a weighted result;
s403: based on the risk index and the weighted result, establishing a risk assessment model by adopting a Bayesian network;
s404: the risk assessment data are imported into a risk assessment model, and the first 10% is defined as high risk, the middle 25% as medium risk and the last 65% as low risk according to the output result of the risk assessment model;
s405: and (5) according to regions with different risk grades, corresponding management measures and prevention strategies are formulated.
Referring to fig. 5, according to regions with different risk levels, corresponding management measures and prevention strategies are formulated, specifically, limit construction, enhance monitoring and provide an early warning system are adopted in the high-risk regions;
taking earthquake preparation measures in the medium risk areas, including disaster education and exercise;
low risk areas focus on the trend and monitoring of risk.
The step of classifying the risk level of a geological disaster into three levels, high, medium and low, has several benefits. Firstly, by determining the applicable risk index and weighting calculation, various potential factors of geological disasters can be comprehensively considered, and the comprehensiveness and accuracy of evaluation are ensured. And secondly, a risk assessment model is established to comprehensively analyze a plurality of indexes, and the correlation and influence of the indexes are considered, so that the reliability and the interpretation of assessment are improved. In addition, the risk level can be quantitatively represented by demarcating the risk level of the geological disaster, so that a decision maker can intuitively know the risk conditions of different areas, and can pertinently formulate management measures and prevention strategies. Through personalized management, the high-risk areas can be strictly limited to build and strengthen monitoring and early warning measures, the medium-risk areas can strengthen disaster education and exercise, and the low-risk areas can monitor the risk change trend in a key way. In summary, the steps of classifying the risk level of the geological disaster into three levels of high, medium and low effectively improve the accuracy and operability of risk assessment, provide scientific basis for decision makers, reduce the risk and loss caused by the geological disaster, and achieve the purposes of effective disaster management and prevention control.
Referring to fig. 6, the steps of performing comprehensive evaluation and optimization of the model by using the newly collected data are specifically as follows:
s501: storing real-time monitoring data and subsequent disaster event data as an updated data set;
s502: applying the updated data set to the already running prediction model to obtain a new prediction result;
s503: comparing and verifying the new prediction result with the actual observation value, and adopting root mean square error as an evaluation index to evaluate the accuracy and stability of the prediction model to obtain a comprehensive evaluation result;
s504: and optimizing the prediction model based on the comprehensive evaluation result.
Referring to fig. 6, based on the comprehensive evaluation result, the steps for optimizing the prediction model specifically include:
adjusting parameters of the prediction model, and searching for an optimal parameter combination by using grid searching and random searching technologies so as to minimize a prediction error of the prediction model on an updated data set;
performing feature engineering on the updated data set, optimizing feature representation by adopting feature selection, feature transformation and feature combination, and improving the prediction capability of the prediction model on geological disasters to serve as an optimized prediction model;
and carrying out a new round of risk assessment and prediction based on the optimized prediction model, comparing with an actual observed value, and further iterating and optimizing the model.
First, the benefit of storing real-time monitoring data and disaster event data as updated data sets is to preserve the timeliness of the data. By collecting and storing the real-time monitoring data and the subsequent disaster event data, the change condition of the geological disaster can be reflected in time, and the data used by the model is ensured to be up to date. Second, the benefit of applying the updated data set to the already running prediction model is to improve prediction accuracy. By using the latest data, the model can better capture the dynamic change of geological disasters, and the accuracy and reliability of prediction are improved. Next, the benefit of comparing and validating the new predictions against the actual observations is to evaluate the model performance. By comparing the difference between the prediction result and the actual observation value, the accuracy and the stability of the model can be evaluated, and a basis is provided for the follow-up model optimization. The benefit of optimizing the predictive model based on the comprehensive evaluation results is improved model performance. By adjusting parameters of the model and performing feature engineering, the prediction capability and accuracy of the model can be further improved, so that the model better accords with the change of an actual observation value.
Referring to fig. 7, the steps of collecting feedback from various aspects to further improve accuracy of the prediction model by establishing a public participation and information sharing mechanism are specifically as follows:
s601: establishing an online platform, and setting up an online investigation tool as a data collection and integration mechanism;
s602: the public provides feedback opinion and knowledge through an online investigation tool;
s603: integrating and summarizing based on feedback opinion and knowledge, cleaning data, removing abnormal values and repeated data, and obtaining a feedback data set;
s604: based on the feedback data set, using statistical methods and text mining techniques to find patterns, trends and key problems in the feedback data set as available data;
s605: based on the improvement of the available data applied to the prediction model, the data set is updated and revised to reflect the actual situation for the feedback of the input data of the model, and the model parameters or algorithms are optimized and adjusted according to the analysis result.
First, through multiple perspectives and knowledge sharing, professionals and the public in each field can participate in common, share their experiences and perspectives, and thus provide more comprehensive data and insight. Secondly, the public participation enriches the data sources and the data quality, including field observation, professional experience and local knowledge, which can increase the input data of the model and improve the accuracy and the reliability of the model. Third, by analyzing the collected feedback data, patterns, trends, and key problems in the data can be discovered, which provides valuable guidance for further predictions and decisions. In addition, by applying feedback data to optimize and refine the predictive model, including parameter adjustments, algorithm improvements, and dataset updates, performance and adaptability of the model may be improved. In addition, the public participation and information sharing mechanism is beneficial to establishing the trust of the public, enhances the risk management capability of the society, and improves the awareness and coping capability of the public on geological disaster risks. In summary, establishing public participation and information sharing mechanisms provides important support for effectively treating geological disasters through the advantages of multiple views, enriching data sources, discovering key problems, model optimization, social risk management and the like.
Working principle: the step of collecting data includes geological exploration, geological survey, remote sensing technology, ground measurement, meteorological observation, soil sampling, laboratory testing, water level measurement, hydrological observation, seismic observation and slope monitoring. These data cover parameters in terms of geologic structure, topography, weather, soil, hydrology, and earthquakes. Based on the collected parameter data, the step of constructing a predictive model using machine learning includes a combination of feature selection, dataset partitioning, multiple linear regression, and decision trees. The feature selection is performed by calculating pearson correlation coefficients, and parameters with strong correlation with geological disasters are selected as participation data sets. The data set is divided into a training set and a test set for training and evaluation of the model. And combining multiple linear regression with a decision tree, establishing a prediction model, and describing the relationship between parameters and geological disasters. The step of utilizing the internet of things technology and the sensor to conduct real-time monitoring comprises the steps of arranging a sensor network, collecting monitoring data in real time, transmitting the data to a data center through the internet of things technology, and feeding the monitoring data back to a prediction model through a stream processing technology, so that real-time updated risk assessment data is achieved. Based on risk assessment data, the risk grade of geological disasters is divided into three grades, namely high, medium and low, a hierarchical analysis method is used for weighting risk indexes, a Bayesian network is adopted for establishing a risk assessment model, and management measures and prevention strategies are formulated for areas with different risk grades according to the output results of the model. After the predictive model is run for a period of time, the model is comprehensively evaluated and optimized by the newly collected data. And storing the real-time monitoring data and the disaster event data as updated data sets, applying the updated data sets to the running prediction model and obtaining a new prediction result, and evaluating the accuracy and stability of the model by comparing and verifying the updated data sets with the actual observation value and adopting root mean square error so as to optimize the prediction model. And collecting feedback opinions and knowledge of the public by establishing a public participation and information sharing mechanism, integrating and summarizing, cleaning data and acquiring a feedback data set. Patterns, trends and key problems in the dataset are discovered by statistical methods and text mining techniques, and the available data is applied to improvements in the predictive model, including updating and modifying the dataset, optimizing parameters or algorithms of the model. In this way, the accuracy of the predictive model is further improved.
The present invention is not limited to the above embodiments, and any equivalent embodiments which can be changed or modified by the technical disclosure described above can be applied to other fields, but any simple modification, equivalent changes and modification made to the above embodiments according to the technical matter of the present invention will still fall within the scope of the technical disclosure.
Claims (7)
1. The risk evaluation geological disaster early warning method based on multi-method cooperation is characterized by comprising the following steps of:
collecting various parameter data related to a geological disaster;
based on the collected parameter data, constructing a prediction model by utilizing machine learning, and identifying risk factors possibly causing geological disasters;
real-time monitoring is carried out by utilizing the Internet of things technology and a sensor, real-time monitoring data are fed back to the prediction model, and risk assessment data are updated in real time;
based on the risk assessment data, the risk grade of the geological disaster is divided into three grades, namely a high grade, a medium grade and a low grade, so that effective management of areas with different risk grades is realized;
after the predictive model operates for a specified time period, comprehensively evaluating and optimizing the model by utilizing the newly collected data;
the feedback of all aspects is collected by establishing a public participation and information sharing mechanism so as to further improve the accuracy of the prediction model;
the step of constructing a prediction model by machine learning based on the collected parameter data specifically comprises the following steps:
based on the collected parameter data, selecting characteristics, and selecting a participation data set;
dividing the participation data set into a training set and a testing set;
selecting a multiple linear regression combined decision tree to construct the prediction model;
based on the training set, training and parameter tuning are carried out on the prediction model, and a trained prediction model is obtained;
evaluating the trained predictive model based on the test set;
the feature selection specifically means that the pearson correlation coefficient between each feature and the geological disaster is calculated, the absolute value of the pearson correlation coefficient is closer to 1, the linear relation between two variables is stronger, the parameter data are arranged in a descending order based on the pearson correlation coefficient, and the data positioned at the upper 30% are selected as the participation data set;
the participation data set is divided into a training set and a test set, wherein 70% of data in the participation data set is used as the training set for training and parameter tuning, and the rest 30% of data is used as the test set for evaluating the performance and generalization capability of the model;
the multiple linear regression specifically refers to establishing a linear regression model by using the relation between the characteristics of the selected participation data set and the geological disasters, and describing the quantitative relation between the parameters and the geological disasters;
the decision tree specifically refers to that layering division among a plurality of parameters is carried out based on a threshold value or information gain of the characteristics, and rules and risk factors of geological disasters are found;
the step of classifying the risk class of the geological disaster into three classes of high, medium and low based on the risk assessment data comprises the following steps:
confirming risk indexes including terrain gradient and elevation, geological conditions, rainfall and rainfall intensity, temperature and meteorological conditions, groundwater level and hydrologic conditions, soil type and stability and geological disasters;
weighting each risk index by using an analytic hierarchy process to obtain a weighted result;
based on the risk index and the weighted result, establishing a risk assessment model by adopting a Bayesian network;
the risk assessment data are imported into a risk assessment model, and the first 10% is defined as high risk, the middle 25% as medium risk and the last 65% as low risk according to the output result of the risk assessment model;
and (5) according to regions with different risk grades, corresponding management measures and prevention strategies are formulated.
2. The method for early warning a geological disaster based on risk assessment by multi-method cooperation according to claim 1, wherein the step of collecting various parameter data related to the geological disaster is specifically as follows:
obtaining geological structure parameter data, including stratum structures, lithology distribution and fracture zones, through geological exploration and geological investigation;
the method comprises the steps of obtaining topographic and geomorphic parameter data, including ground surface elevation, gradient, slope direction and topographic curvature, by utilizing a remote sensing technology and ground measurement;
collecting weather parameter data including rainfall, snowfall, air temperature, humidity and wind speed through a weather observation station, weather satellites and weather radars;
soil parameter data including soil type, water content and permeability are obtained through soil sampling and laboratory testing;
collecting hydrological parameter data including groundwater level, river level and flood history through a water level measuring instrument and a hydrological observation station;
seismic monitoring data are acquired through a seismic observation station, ground temperature data are acquired through a ground temperature measuring device, and slope displacement data are acquired through slope monitoring equipment.
3. The method for early warning geological disasters based on multi-method cooperation risk evaluation according to claim 1, wherein the steps of real-time monitoring by using internet of things technology and sensors, feeding real-time monitoring data back to the prediction model, and real-time updating risk evaluation data are specifically as follows:
arranging a sensor network in a detection area, and connecting the sensor with the network through the technology of the Internet of things, wherein the sensor comprises an earthquake sensor, a displacement sensor and a pressure sensor;
the sensor collects monitoring data in real time and transmits the monitoring data to a data center through the internet of things technology;
the data center feeds the monitoring data back to the prediction model based on a stream processing technology;
and updating the prediction model by using real-time monitoring data, carrying out real-time risk assessment according to the output of the prediction model, and updating the risk assessment data.
4. The method for early warning geological disasters based on multi-method cooperation risk evaluation according to claim 1, wherein the corresponding management measures and prevention strategies are formulated according to regions with different risk grades, and the high-risk regions adopt limit construction, enhanced monitoring and early warning systems;
taking earthquake preparation measures in the medium risk areas, including disaster education and exercise;
low risk areas focus on the trend and monitoring of risk.
5. The method for early warning geological disasters based on multi-method collaboration risk assessment according to claim 1, wherein the steps of comprehensively evaluating and optimizing the model by using the newly collected data are specifically as follows:
storing the real-time monitoring data and the subsequent disaster event data as an updated data set;
applying the updated data set to the prediction model that has been run to obtain a new prediction result;
comparing and verifying the new prediction result with the actual observation value, and adopting root mean square error as an evaluation index to evaluate the accuracy and stability of the prediction model to obtain a comprehensive evaluation result;
and optimizing the prediction model based on the comprehensive evaluation result.
6. The method for early warning a geological disaster based on risk assessment by multi-method cooperation according to claim 5, wherein the step of optimizing the prediction model based on the comprehensive assessment result is specifically as follows:
adjusting parameters of the prediction model, and searching for an optimal parameter combination by using grid searching and random searching technologies so as to minimize a prediction error of the prediction model on an updated data set;
performing feature engineering on the updated data set, optimizing feature representation by adopting feature selection, feature transformation and feature combination, and improving the prediction capacity of the prediction model on geological disasters to serve as an optimized prediction model;
and carrying out a new round of risk assessment and prediction based on the optimized prediction model, comparing with an actual observed value, and further iterating and optimizing the model.
7. The method for early warning geological disasters based on multi-method collaboration risk assessment according to claim 1, wherein the step of collecting feedback from all aspects by establishing a public participation and information sharing mechanism to further improve accuracy of a prediction model is specifically as follows:
establishing an online platform, and setting up an online investigation tool as a data collection and integration mechanism;
the public provides feedback opinion and knowledge through the online survey tool;
integrating and summarizing based on the feedback opinion and knowledge, cleaning data, removing abnormal values and repeated data, and obtaining a feedback data set;
based on the feedback data set, using a statistical method and a text mining technique to find patterns, trends and key problems in the feedback data set as available data;
based on the improvement of the available data applied to the prediction model, the data set is updated and corrected for feedback of model input data to reflect actual conditions, and optimization and adjustment are performed according to analysis results for feedback of model parameters or algorithms.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311288116.5A CN117037432B (en) | 2023-10-08 | 2023-10-08 | Risk evaluation geological disaster early warning method based on multi-method cooperation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311288116.5A CN117037432B (en) | 2023-10-08 | 2023-10-08 | Risk evaluation geological disaster early warning method based on multi-method cooperation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117037432A CN117037432A (en) | 2023-11-10 |
| CN117037432B true CN117037432B (en) | 2023-12-19 |
Family
ID=88630310
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311288116.5A Active CN117037432B (en) | 2023-10-08 | 2023-10-08 | Risk evaluation geological disaster early warning method based on multi-method cooperation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117037432B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117593653A (en) * | 2024-01-19 | 2024-02-23 | 山东元鸿勘测规划设计有限公司 | Geological disaster early warning method based on remote sensing monitoring |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102651053A (en) * | 2012-04-01 | 2012-08-29 | 中国科学院地理科学与资源研究所 | Evaluation method of debris flow disaster risk based on Bayesian network model |
| CN109003422A (en) * | 2018-08-02 | 2018-12-14 | 北京大学深圳研究生院 | Monitoring data processing method and landslide forecasting procedure for landslide |
| CN112735094A (en) * | 2020-12-17 | 2021-04-30 | 中国地质环境监测院 | Geological disaster prediction method and device based on machine learning and electronic equipment |
| CN113470316A (en) * | 2021-05-24 | 2021-10-01 | 电子科技大学 | Debris flow monitoring and early warning method based on self-coding single classification model |
| CN115394052A (en) * | 2022-08-30 | 2022-11-25 | 重庆地质矿产研究院 | Method for obtaining geological disaster early warning key parameter prediction value based on machine learning |
| CN116090696A (en) * | 2022-12-01 | 2023-05-09 | 电子科技大学 | Landslide geological disaster risk classification prediction method suitable for mountain railway line |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2022256517A1 (en) * | 2021-04-16 | 2023-11-09 | Strong Force Vcn Portfolio 2019, Llc | Systems, methods, kits, and apparatuses for digital product network systems and biology-based value chain networks |
-
2023
- 2023-10-08 CN CN202311288116.5A patent/CN117037432B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102651053A (en) * | 2012-04-01 | 2012-08-29 | 中国科学院地理科学与资源研究所 | Evaluation method of debris flow disaster risk based on Bayesian network model |
| CN109003422A (en) * | 2018-08-02 | 2018-12-14 | 北京大学深圳研究生院 | Monitoring data processing method and landslide forecasting procedure for landslide |
| CN112735094A (en) * | 2020-12-17 | 2021-04-30 | 中国地质环境监测院 | Geological disaster prediction method and device based on machine learning and electronic equipment |
| CN113470316A (en) * | 2021-05-24 | 2021-10-01 | 电子科技大学 | Debris flow monitoring and early warning method based on self-coding single classification model |
| CN115394052A (en) * | 2022-08-30 | 2022-11-25 | 重庆地质矿产研究院 | Method for obtaining geological disaster early warning key parameter prediction value based on machine learning |
| CN116090696A (en) * | 2022-12-01 | 2023-05-09 | 电子科技大学 | Landslide geological disaster risk classification prediction method suitable for mountain railway line |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117037432A (en) | 2023-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Herath et al. | Hydrologically informed machine learning for rainfall–runoff modelling: towards distributed modelling | |
| Greve et al. | Quantifying the ability of environmental parameters to predict soil texture fractions using regression-tree model with GIS and LIDAR data: The case study of Denmark | |
| Segoni et al. | Landslides triggered by rainfall: A semi-automated procedure to define consistent intensity–duration thresholds | |
| Thirumalaivasan et al. | AHP-DRASTIC: software for specific aquifer vulnerability assessment using DRASTIC model and GIS | |
| CN117037432B (en) | Risk evaluation geological disaster early warning method based on multi-method cooperation | |
| WO2013009160A2 (en) | A geometric method for predicting landslide disaste | |
| CN109886456B (en) | Debris flow prediction method based on PCA and mixed kernel function LSSVR | |
| Mohammadi et al. | Estimation of forest stand volume, tree density and biodiversity using Landsat ETM+ Data, comparison of linear and regression tree analyses | |
| Chebbi et al. | Optimal extension of rain gauge monitoring network for rainfall intensity and erosivity index interpolation | |
| Sujatha et al. | Landslide susceptibility mapping of Tevankarai Ar sub-watershed, Kodaikkanal taluk, India, using weighted similar choice fuzzy model | |
| CN115660407A (en) | Training method and training device for pollution site risk grade prediction model | |
| CN107957982B (en) | Method and system for rapidly evaluating easiness of post-earthquake secondary geological disasters | |
| Smith et al. | Spatial variability and uncertainty of soil nitrogen across the conterminous United States at different depths | |
| CN116110210B (en) | Data-driven landslide hazard auxiliary decision-making method in complex environment | |
| Lu et al. | Uncertainty quantification of machine learning models to improve streamflow prediction under changing climate and environmental conditions | |
| Azarnivand et al. | Analysing the uncertainty of modelling hydrologic states of AWRA-L–understanding impacts from | |
| CN108154263B (en) | Monitoring and predicting method for natural water resource | |
| CN116738192A (en) | Digital twinning-based security data evaluation method and system | |
| Wang et al. | Early warning of debris flow using optimized self-organizing feature mapping network | |
| CN116402951A (en) | Digital twinning-based side slope safety management method and system | |
| CN116090696A (en) | Landslide geological disaster risk classification prediction method suitable for mountain railway line | |
| Park et al. | Settlement prediction in a vertical drainage-installed soft clay deposit using the genetic algorithm (GA) back-analysis | |
| CN112836842A (en) | Watershed water environment quality prediction method and system based on source-sink risk analysis | |
| County | Monitoring for adaptive management: Status and trends of aquatic and riparian habitats in the Lake Washington/Cedar/Sammamish Watershed (WRIA 8) | |
| CN117436823B (en) | Natural resource right-determining registration management method and system based on big data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |